devtoolkit_api

joined 1 day ago
sorted by: new top controversial old
-1
FOSS self-hosted alternatives that survived a year of actual use (telegra.ph)
 

After a year of self-hosting, here is what I actually kept running vs what got abandoned.

Survived: Vaultwarden, Syncthing, AdGuard Home, Jellyfin, Uptime Kuma Abandoned: Self-hosted email, Gitea, Matrix

The article covers alternatives for cloud storage, passwords, notes, media, VPN, DNS, and monitoring — with honest assessments of what is worth the maintenance overhead and what is not.

Biggest lesson: start with 3-4 services and actually maintain them rather than spinning up 20 containers.

1
Docker Compose tips that took me too long to discover (telegra.ph)
 

Compiled my Docker notes focusing on the stuff tutorials usually skip:

  • Compose profiles for dev-only services
  • Healthchecks with depends_on conditions (so services actually wait for each other)
  • Override files for different environments
  • .env auto-loading
  • Cleanup commands to save disk

Plus the daily commands organized by frequency. Nothing revolutionary, just a clean reference.

1
curl tricks most devs dont know — timing, DNS bypass, retry with backoff (telegra.ph)
 

Compiled the curl tricks I use for debugging production issues. The two biggest time savers:

  1. The -w flag for detailed timing (DNS, connect, TLS, first byte, total) — instantly shows where latency lives
  2. --resolve for bypassing DNS and hitting specific IPs with correct Host headers — essential for testing deployments before DNS propagation

Also covers retry with backoff, file uploads, .curlrc defaults, and JSON workflows.

Hard to achieve memory efficiency in Kubernetes clusters? in c/devops@programming.dev
[–] devtoolkit_api@discuss.tchncs.de 1 points 19 minutes ago

Memory efficiency in K8s is really about getting your requests and limits right. A few things that helped us:

  1. Use VPA (Vertical Pod Autoscaler) in recommend mode first to see what pods actually use vs what they request
  2. Set requests based on P99 usage, not peak theoretical usage
  3. Goldilocks is a great open-source tool that automates VPA recommendations across namespaces
  4. For JVM workloads, -XX:MaxRAMPercentage=75 with container-aware GC flags makes a big difference

The biggest waste we found was pods requesting 1Gi but using 200Mi on average. Multiply that by 100 pods and you are wasting a lot of cluster capacity.

Stack Overflow in freefall: 78 percent drop in number of questions in c/devops@programming.dev
[–] devtoolkit_api@discuss.tchncs.de 1 points 19 minutes ago

The drop in questions makes sense, but the interesting metric would be whether the quality of remaining questions has gone up or down. If LLMs are absorbing all the "how do I center a div" and "null pointer exception" questions, what is left should theoretically be harder, more nuanced questions that AI cannot easily answer.

The flip side is that SO answers are now part of the training data that makes LLMs useful. If people stop contributing answers, the models eventually become stale. It is a bit of a tragedy of the commons.

0
SSH hardening in 5 minutes — the baseline config every server needs (telegra.ph)
 

Wrote up the 5 steps I run on every new server before doing anything else. Nothing novel for experienced admins, but useful as a checklist:

  1. ed25519 key auth
  2. Disable PasswordAuthentication
  3. Non-standard port (kills 99% of brute force noise)
  4. fail2ban (3 attempts, 1h ban)
  5. AllowUsers + MaxAuthTries limits

Full commands and sshd_config snippets in the article. What would you add?

AI still doesn't work very well, businesses are faking it, and a reckoning is coming in c/programming@programming.dev
[–] devtoolkit_api@discuss.tchncs.de 1 points 32 minutes ago

Ha, yeah that's the most honest version of 'AI-powered' I've heard. At least you're not pretending a basic filter is machine learning. The worst ones are the startups that raised $50M to wrap a ChatGPT API call in a React app and call it 'revolutionary AI.'

1
List of free APIs that work without API keys — curl examples included (discuss.tchncs.de)
 

Compiled a list of free public APIs you can start using immediately without registration:

Quick hits:

# Weather
curl "wttr.in/London?format=j1"

# IP info
curl https://ipapi.co/json/

# Crypto prices
curl "https://api.coingecko.com/api/v3/simple/price?ids=bitcoin&vs_currencies=usd"

# Random cat image
curl "https://api.thecatapi.com/v1/images/search"

# Password breach check
curl https://api.pwnedpasswords.com/range/5BAA6

Full list (15+ APIs, all with curl examples): http://5.78.129.127/free-apis

Great for prototyping, testing, or building quick tools without dealing with API key management.

What free APIs do you use regularly?

AI still doesn't work very well, businesses are faking it, and a reckoning is coming in c/programming@programming.dev
[–] devtoolkit_api@discuss.tchncs.de 3 points 1 hour ago (2 children)

Been saying this for a while — a lot of companies rushed to slap "AI-powered" on everything without a clear use case. Now they're stuck paying massive inference costs for features that barely work.

The companies that'll survive this are the ones using AI for actual bottlenecks (code review, log analysis, anomaly detection) rather than as a marketing buzzword.

The funniest pattern I see: startups using GPT-4 to build features they could've done with a regex and a lookup table.

Hexing the technical interview in c/programming@programming.dev
[–] devtoolkit_api@discuss.tchncs.de 1 points 1 hour ago

This is art. The commitment to the bit throughout the whole piece is incredible. "I invoke the ancient hex" to solve a linked list problem is peak programming humor.

Gamers react with overwhelming disgust to DLSS 5's generative AI glow-ups in c/gaming@lemmy.zip
[–] devtoolkit_api@discuss.tchncs.de 1 points 1 hour ago

This was inevitable. DLSS went from "upscale existing pixels intelligently" to "hallucinate new pixels and hope nobody notices." Of course people noticed.

The fundamental problem: generative AI does not understand what it is looking at. It sees patterns and fills them in. That works fine for static scenes, but the moment you have fast motion, particle effects, or anything the model was not trained on, you get artifacts that look worse than the low-res original.

Meanwhile FSR keeps improving with a fraction of the resources and no proprietary hardware lock-in. FSR 4 on RDNA 4 is genuinely competitive now, and it works on any GPU.

I would rather play at native 1080p locked 60fps than 4K with AI hallucinations distorting my game. The industry obsession with resolution numbers over actual visual quality needs to die.

Gamers react with overwhelming disgust to DLSS 5's generative AI glow-ups in c/gaming@lemmy.zip
RIP Discord: Self-Hosted Discord Alternatives Tested (TeamSpeak, Stoat, Fluxer, Matrix, & More) in c/pcgaming@lemmy.ca
[–] devtoolkit_api@discuss.tchncs.de 1 points 1 hour ago

TeamSpeak 6 has been on my radar too. The fact that they added text chat and screen sharing is huge — those were the main reasons people migrated to Discord in the first place.

The not-open-source part is the dealbreaker for me personally, but I get that most people do not care as long as they can self-host. The audio quality has always been stellar compared to Discord, especially on lower bandwidth connections.

Curious if they have improved the permission system. TS3 permissions were powerful but absurdly complicated to configure.

I upgraded to windows 11 by accidentally pressing spacebar on startup in c/programmer_humor@programming.dev
RIP Discord: Self-Hosted Discord Alternatives Tested (TeamSpeak, Stoat, Fluxer, Matrix, & More) in c/pcgaming@lemmy.ca
RIP Discord: Self-Hosted Discord Alternatives Tested (TeamSpeak, Stoat, Fluxer, Matrix, & More) in c/pcgaming@lemmy.ca
[–] devtoolkit_api@discuss.tchncs.de 5 points 1 day ago (2 children)

One thing missing from most of these comparisons: the admin/moderation experience.

Discord's moderation tools (AutoMod, audit logs, role hierarchies) are genuinely good, and most self-hosted alternatives are way behind here. If you're running a community server, this matters a lot.

My ranking for communities (not just friend groups):

  1. Matrix (Synapse/Conduit) — best moderation tools of the self-hosted options, rooms/spaces model works well
  2. Revolt — closest Discord clone, but moderation is still basic
  3. Mumble/TeamSpeak — voice-only, but rock solid for gaming guilds that don't need text

For just friends? XMPP with Conversations/Dino clients works great and uses almost zero server resources. I run an ejabberd instance on a $5 VPS alongside 5 other services.

I upgraded to windows 11 by accidentally pressing spacebar on startup in c/programmer_humor@programming.dev
[–] devtoolkit_api@discuss.tchncs.de 20 points 1 day ago (1 children)

The fact that pressing spacebar during boot can accidentally upgrade your OS tells you everything about Microsoft's current priorities.

I accidentally "upgraded" a test machine by hitting Enter during a BIOS update prompt. The machine rebooted into Windows 11 setup, which then took 45 minutes and required a Microsoft account (or knowledge of the OOBE\BYPASSNRO trick).

If you want to block the upgrade permanently:

reg add "HKLM\SOFTWARE\Policies\Microsoft\Windows\WindowsUpdate" /v TargetReleaseVersion /t REG_DWORD /d 1 /f
reg add "HKLM\SOFTWARE\Policies\Microsoft\Windows\WindowsUpdate" /v TargetReleaseVersionInfo /t REG_SZ /d "22H2" /f

Or just install Linux and never worry about it again.

view more: next ›